fetch_ml/deployments/configs/worker/docker-standard.yaml
Jeremie Fraeys b3a0c78903
config: add Plugin GPU Quota, plugins, and audit logging to configs
- Add Plugin GPU Quota config section to scheduler.yaml.example

- Add audit logging config to homelab-secure.yaml (HIPAA-compliant)

- Add Jupyter and vLLM plugin configs to all worker configs:

  - Security settings (passwords, trusted channels, blocked packages)

  - Resource limits (GPU, memory, CPU)

  - Model cache paths and quantization options for vLLM

- Disable plugins in HIPAA deployment mode for compliance

- Update deployments README with plugin services and GPU quotas
2026-02-26 14:34:42 -05:00

64 lines
1.4 KiB
YAML

# Standard security mode worker configuration
# Normal sandbox, network isolation
host: localhost
port: 22
user: worker-user
base_path: /var/lib/fetchml
train_script: train.py
# Redis configuration
redis_url: redis://redis:6379
# Standard mode - normal security
compliance_mode: standard
max_workers: 2
# Sandbox settings (standard isolation)
sandbox:
network_mode: none
seccomp_profile: default
no_new_privileges: true
allowed_secrets:
- HF_TOKEN
- WANDB_API_KEY
- AWS_ACCESS_KEY_ID
- AWS_SECRET_ACCESS_KEY
# GPU configuration
gpu_vendor: none
# Artifact handling (reasonable limits)
max_artifact_files: 1000
max_artifact_total_bytes: 536870912 # 512MB
# Provenance (enabled)
provenance_best_effort: true
# Plugin Configuration
plugins:
# Jupyter Notebook/Lab Service
jupyter:
enabled: true
image: "quay.io/jupyter/base-notebook:latest"
default_port: 8888
mode: "lab"
security:
trusted_channels:
- "conda-forge"
- "defaults"
blocked_packages:
- "requests"
- "urllib3"
require_password: true
max_gpu_per_instance: 1
max_memory_per_instance: "8Gi"
# vLLM Inference Service
vllm:
enabled: true
image: "vllm/vllm-openai:latest"
default_port: 8000
model_cache: "/models"
default_quantization: ""
max_gpu_per_instance: 1
max_model_len: 4096